The aim of this continuation project is to characterize how early and mid adolescent neurocognitive functioning, in the context of an emerging drinking career, serves to increase risk for the emergence of alcohol use disorder. 899 participants from an ongoing longitudinal study of at risk children, who had neurocognitive assessments in early adolescence during the prior grant period, will move from late adolescence into their early twenties during the current project. During this crucial developmental transition the heaviest levels of alcohol and other drug use can be expected. The project's conceptual framework integrates three broad functional neural networks theorized to be involved in alcoholism and other drug abuse which will be assessed via performance measures: (a) executive functioning, especially response suppression and working memory, (b) reward response, including anticipation of reward and response to reward, and (c) secondarily, right hemisphere function, with discrimination between spatial and social information processing. Bi-directional behavior-neurocognitive effects will be considered. Neural functioning will also be characterized directly, via fMRI, to examine: (1) individual differences in brain functioning in relation to alcohol/drug use, neurocognitive performance, and earlier neurocognitive weakness; and (2) changes in brain functioning over time with increasing drug use, also assessed in relation to earlier neurocognitive weakness. Tasks involving the two core domains of the proposal, executive functioning and reward response, will be used to activate brain regions of interest. We will characterize differences in neural development in the late teens and again two years thereafter?as the frontal circuitry of interest matures?in 120 selected youth from the larger study. This design will enable determination of whether neural activation patterns are similar, with and without performance decrements, and in relation to alcohol/drug use during the intervening 2 years. The project will be carried out in a well characterized longitudinal sample of male and female children of alcoholics along with an ecologically comparable but nonalcoholic group of controls who have been assessed at three year intervals since early childhood. Neurocognitive assessments will be carried out at 3 year intervals, and the fMRI protocol is to be administered twice.